Acute myeloid leukemia (AML) is an aggressive blood cancer with a poor prognosis, with survival rates in younger patients at approximately 50%, but less than 20% in patients over 60 years old. Additionally, AML stem cells often survive chemotherapy in the protective environment of the bone marrow, causing relapse even in patients with a good initial response to therapy. The current standard of care, chemotherapy or stem cell transplant, has remained largely unchanged for more than three decades. Unfortunately, of the estimated 18,000 Americans diagnosed with AML every year, more than 10,000 will die of their disease, defining an urgent need for better and more targeted therapies. DNA sequencing has identified numerous mutations in AML patients, but as yet this knowledge has not translated into major therapeutic advances. This probably reflects the complexity of the disease and the fact that many of the mutated genes have proven difficult to target therapeutically. To overcome some of the limitations of past efforts, we have taken a function-first approach, where AML cells from patients are screened and thousands of genes are inactivated, one-by-one, so that we can identify those that are critical for AML cell survival. Importantly, our assays are done in the presence of bone marrow stromal cells to mimic the protective environment of the bone marrow. This screen has revealed that a subset of AML samples are dependent on the activity of SIRT5, a multifunctional enzyme that regulates vital metabolic pathways in the mitochondria and cytosol. In these cells, SIRT5 knockdown caused growth inhibition and cell death, while normal, healthy bone marrow cells were unaffected. We propose to use SIRT5 knockdown and metabolic profiling to determine other key survival pathways in AML stem cells with the intention of exploiting these pathways with small molecules. These data will be correlated with somatic mutation screening, gene expression analysis and metabolomics profiling to understand the mechanistic underpinnings of SIRT5 dependence. Our studies have the potential to establish proof of principle that SIRT5 is a therapeutic target in AML, providing a scientific rationale for the development of clinically-viabl SIRT5 inhibitors to treat AML and possibly other types of cancer. Importantly, mice with deletion of SIRT5 are viable and have no major defects, suggesting that SIRT5 inhibitors would be well tolerated.